Evidence Atlas

Research database

How evidence tiers are defined

Level A

Multiple randomised controlled trials with reproducible human data.

Level B

Preliminary human studies, or strong animal models with mechanistic plausibility.

Level C

Mechanistic or early-stage work, often with no human data yet.

Grades are assigned conservatively. Where the evidence is mixed, or where a strong preclinical result is not matched by human data, the lower grade is shown. Tier reflects the strength and replication of evidence, not the size of the effect, and is never signalled by colour alone.

Category
Evidence level

73 of 73 records